Offset Detection Of Grate Trolley’s Side Plate Based On Machine Vision

In recent years,with the rapid development of China’s iron and steel industry,iron ore has been mined in large quantities,and the natural lump ore with high iron grade is decreasing day by day.At present,pellets have become the main raw material for China’s blast furnace ironmaking.Grate-kiln pelletizing process is a widely used pelletizing production method in Chinese iron and steel enterprises,its main equipment consists of grate,rotary kiln and annular cooler.The grate machine is mainly used for drying,preheating and oxidizing pellets,and the output and quality of pellets depend on its working condition directly.Side plate offset is one of grate system faults.If it is not dealt with in time,some accidents will occur and economic losses will be made.At present,the offset detection of the side plate is still mainly by manual observation method,which is time-consuming,labour-wasted and low intelligent.Nowadays,machine vision-based defect detection and fault diagnosis methods have been widely used in various industrial fields.Inspired by related research,the real-time automatic detection was realized to detect the offset fault of grate trolley’s side plate,based on machine vision.The research in this thesis mainly includes the followings: Firstly,through the full investigation of the grate work site in Baogang Group sintering plant,combined with the operation and maintenance specifications of the grate machine,the side plate offset detection system of the grate trolley is designed,and the fault classification and alarm rules of trolley’s side plate offset are formulated.By proposing concepts such as "demarcation line","baseline","low-level offset fault" and "high-level offset fault",to provide a theoretical basis for the implementation of subsequent side plate offset detection work.Then,based on the idea of "converting the detection of the side plate offset into the detection of the straight-line segment offset",using the image line segment detection technology,a side plate offset detection scheme based on Line Segment Detector(LSD)algorithm is proposed.The scheme can effectively reduce the working intensity of on-site workers,with high degree of automation,and good real-time performance.Next,in view of the shortcomings of the side plate offset detection scheme based on LSD algorithm,the deep learning object detection algorithm is used to replace the image line segment detection technology to complete the object detection task,and a side plate offset detection scheme based on You Only Look Once version 4(YOLOv4)is proposed,which further reduces manual intervention,lower the program’s false detection rate and improves the speed of fault detection.Finally,based on Open CV and Py Qt5,the side plate offset detection software system is designed and developed,by using multi-threaded design method.The designed software system can run stably in Windows environment for a long time.In this thesis,machine vision and image processing technology are applied to the pellet production field of iron and steel industry.The data collected and sorted in Baogang Group sintering plant were tested.The results show that the method in this thesis,rather than manually,the real-time automatic detection was realized to detect the offset fault of trolley’s side plate so as to provide a new solution for offset detection of grate trolley’s side plate.